LNG is a groundbreaking laboratory for studies relating brain gene polymorphisms to in vitro function as well as in vivo functional measures. Several of the polymorphism/function associations are congruent with the known molecular neurobiology of the transmitters and proteins and include dopamine transporter genotype/dopamine transporter density in striatum, serotonin transporter genotype/serotonin transporter density in raphe nucleus, COMT genotype/metabolic activity in frontal lobe, and the relationship of COMT to pain/stress and ability to activate endogenous opioid release and metabolic activity in amygdala and other parts of the limbic system. Functional variants are the endgame of positional cloning and are invaluable for selecting the appropriate phenotypes for linkage studies. LNG is screening noncoding sequences for effects on transcription and RNA processing, and evaluating certain coding variants of GPCRs for altered ligand affinity, signal transduction and receptor downregulation. To evaluate effects of sequence variants [including unknown ones] on imprinting or DNA transcription and RNA processing, we created a 5' exonuclease-based method differential allele expression using transcribed SNPs as endogenous reporters. This method was validated and used to detect haplotype effects on expression of COMT alleles previously implicated in behavior. This work raises the possibility of a double-hit for COMT in individuals who more highly express the high activity Val158 allele. Due to neuroadaptation, phenotypic effects of functionally significant alleles may be difficult to discern except in cellular expression systems in which the genetic background on which the various alleles are assayed is identical. Therefore, we transfect allele-specific reporter constructs into a variety of neuronal and non-neuronal cell lines. For example, we detected two missense substitutions in an intracellular loop of the 5HT2A receptor: Ala447Val [allele frequency 0.007] and His452Tyr [allele frequency 0.093. Due to the high frequency of Tyr452 and the expression of authentic 5HT2A in platelets, we were able to compare the functional properties of the His452 and Tyr452 alleles in eight His452/Tyr452 heterozygotes to eight His452/His452 homozygotes matched for sex, age and diagnosis. After 10 mM serotonin, calcium mobilization was reduced in His452/Tyr452 heterozygotes and decay of stimulated intracellular levels was prolonged. These observations were replicated in transiently transfected cells. A 5HT2C Cys23Ser variant we detected has an allele frequency of 0.13. HTR2C is X-linked; therefore, 13% of males are hemizygous for Ser23 and 87% are hemizygous for Cys23. Both alleles were expressed in two highly distinctive cellular environments: cos-7 kidney cells and Xenopus oocytes, and we found that Ser23 allele is constitutively active and therefore downregulated. Other SNPs we detected which alter ligand affinity are OPRM1 Asn40Asp, shown by Bond et al to alter ligand affinity and the DRD2 dopamine polymorphism Ser311Cys, which alters affinity and transduction. The functionality of the serotonin transporter [SLCA4] polymorphism which is associated with anxiety and alters in vitro transcription was pursued. If the mechanism of the SLCA4 linkage to anxiety is to alter transcription, a validating step would be to demonstrate an effect of the polymorphism on serotonin transporter density in human brain, and this was done first with Andreas Heinz and recently with the group at NYU (Oquendo et al; Parsey et al). In controls, the lower transcribing s allele was indeed associated with lower transporter density. Serotonin transporter genotype predicted amygdala metabolic activity to a cognitive fear challenge (Egan et al) and to molecular and behavioral responses to tryptophan depletion (Neumeister et al; Finger et al), providing a bridge between the functionality of this variant and complex behavior. A rare serotonin transporter amino acid substitution was detected in two families. This allele leads to higher transporter function and to behavioral problems including severe,treatment resistant OCD. The serotonin transporter promoter locus itself had been the object of over 300 genetic studies, but we discovered that a common allele at this locus was functional, so that the locus was functionally triallelic instead of biallelic. This A>G SNP in the first of two extra repeats defining the L allele alters mRNA expression by creating an AP2 binding site. Using this new information, we linked the high activity HTTLPR genotype to OCD in both an NIMH case/control dataset (D. Murphy) and replicated this finding in a transmission equilibrium dataset from Toronto (J. Kennedy). The common (allele frequency=0.42), functional COMT Val158Met was linked to frontal cortical function in datasets with widely differing baseline performances. Gene dosage relationship to frontal lobe cognitive performance on was seen in controls, head injured patients, schizophrenics, and siblings of schizophrenics. COMT genotype was linked to frontal cortical coherence. These studies were followed by metabolic brain imaging (BOLD, MRI) during a working memory task which accesses functions of the frontal lobe. As predicted, the Val158 allele correlated with cortical inefficiency during the working memory task. Recently, we have linked Met158 to effects which may be evolutionarily counterbalancing: namely, anxiety in women and diminished pain/stress tolerance (as we originally reported in Zubieta et al and replicated in Diatchenko et al), ability of the endogenous opioid system to activate after a painful stimulus (Zubieta at al)and emotional processing (Smolka et al). BDNF Val66Met was found to impair trafficking and secretion of the peptide, and was linked to Hippocampal metabolic activity and volume (Egan et al; Szesko et al. Using differential allele expression in postmortem human brain, functional haplotypes were defined for both MPDZ and NPY and we discovered linkage of NPY (a peptide involved in stress/anxiety response) to anxiety. Also in the pain genetics domain, we helped show (with collaborators at Harvard and Mitchell Max/Inna Belfer at NIDCR, that GTP cyclohydrolase is a pain-regulated gene in the CNS and has a functional polymorphism linked to chronic clinical pain (Tegeder et al, Nature Medicine, In Press).